The invention discloses a traffic speed prediction method based on a time-space dependency enhancement network, and the method employs the time dependency enhancement network to learn the time correlation in a traffic speed data sequence, and employs the seasonal tendency decomposition to process the seasonality and tendency in the traffic speed data sequence, thereby achieving the prediction of the traffic speed. GRU-Attention is introduced to learn a time dependency relationship in a component obtained through seasonal tendency decomposition; secondly, learning a spatial dependency relationship in the traffic data sequence by adopting a graph attention network; and the output of the time dependence enhancement network and the output of the spatial dependence enhancement network formed by the graph attention network are spliced as the input of the full connection layer, so that a final prediction result is obtained, and the traffic speed prediction precision is improved.
本发明公开了一种基于时空依赖增强网络的交通速度预测方法,该方法采用时间依赖增强网络学习交通速度数据序列中的时间相关性,其中,利用季节趋势性分解来处理交通速度数据序列中的季节性和趋势性,引入GRU‑Attention来学习经过季节趋势性分解得到的分量中的时间依赖关系;其次,采用图注意力网络来学习交通数据序列中的空间依赖关系;并将时间依赖增强网络的输出与图注意力网络构成的空间依赖增强网络的输出进行拼接作为全连接层的输入,从而得到最终的预测结果,提高了交通速度预测的精度。
Traffic speed prediction method based on space-time dependency enhancement network
一种基于时空依赖增强网络的交通速度预测方法
2024-05-28
Patent
Electronic Resource
Chinese
Traffic speed prediction method based on multi-space-time diagram fusion convolutional network
European Patent Office | 2023
|Traffic flow speed prediction method based on attention space-time diagram convolutional network
European Patent Office | 2023
|European Patent Office | 2023
|Traffic speed prediction method based on multi-spatial scale space-time Transform
European Patent Office | 2023
|Traffic flow prediction method based on space-time neural network
European Patent Office | 2024
|